fenn brlcad: i converted the 19MB ApplicationDevelopment file to html, so 
that others wouldn't have to download a 19MB file and find out there was 
only about 20K of text
fenn http://fenn.dyndns.org/pub/ApplicationDevelopment.html
* tegtmeye (n=tegtmeye@pool-70-17-225-27.balt.east.verizon.net) has joined 
#brlcad
brlcad hmm
brlcad ~ping fenn.dydns.org
ibot pong fenn.dydns.org
brlcad fenn: thanks, I'll post it
brlcad (and agree)
brlcad it was a presentation that was just provided for convenience since 
it's a fairly frequent request
brlcad fenn: mind you that presentation is fairly specific to writing 
analysis-style applications
fenn right
brlcad if you are writing out geometry, you don't really need to be 
concerned with most of that
fenn how do i test intersection of two solids?
fenn do a boolean intersection and see if anything is left?
brlcad ah, we call those overlaps
brlcad confused me there for a sec
fenn er sorry, i'm talking about collision detection
brlcad boolean intersections inherintly don't "overlap"
brlcad right
fenn havent slept yet :)
brlcad me either ;)
brlcad almost time for midday coffee
brlcad for testing intersection of two implicitly defined objects, you 
have to use ray introspection
fenn .
fenn oops
brlcad just mathematically due to the nature of implicits
brlcad now if you convert it to an explicit model, there are other 
techniques available
brlcad but brl-cad deals best with implicits currently
fenn what is the difference between explicit and implicit geometry?
brlcad implicit geometry has no defined boundary, it's a mathematical 
model
brlcad numerically defined by mathematical geometric shapes
fenn like a sphere
brlcad e.g. an ellipsoid defined by a point and major and minor radius
brlcad take that same object and represent it differently
fenn is a bot explicit?
brlcad say, as a bunch of triangles on the surface
fenn ok
brlcad and you have explicit
brlcad or a set of surface splines, also explicit
brlcad very different problems determining whether i'm "inside" a sphere 
comprised of a bunch of triangles, or one that's implicit
brlcad for the implicit form, it might be a simple distance test for a 
sphere, if distance to center is < radius then I'm inside
fenn can i apply a n*m transformation matrix and still end up with 
implicit geometry?
brlcad for explicit, if all I have are the triangles, I'm performing 
planar side checks
brlcad sure
fenn (for that matter can i apply a n*m matrix at all?)
brlcad you can apply all the same rigid, non-rigid transformations
brlcad brl-cad has a couple primitives that are not well behaved for 
certain non-rigid transformations (because mathematically they are not 
well defined for them) like trying to skew o torus for example